start
Pass
Audited by Gen Agent Trust Hub on Jul 26, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill scans project-level files such as package.json, DESIGN.md, and .ui-craft/brief.md to detect project state. This creates a surface where external, untrusted data is ingested into the agent's context.
- Ingestion points: package.json, DESIGN.md, .ui-craft/brief.md, .ui-craft/spec.md, design-tokens.json.
- Boundary markers: No explicit delimiters are used to separate external file content from agent instructions.
- Capability inventory: While this specific skill is read-only, it routes users to other commands like /brief, /tokens, and /craft which possess file-writing and code-modification capabilities.
- Sanitization: The skill does not perform sanitization or validation of the content read from the project files.
- [EXTERNAL_DOWNLOADS]: The skill recommends the use of npx ui-craft-detect for continuous integration (CI) environments. This involves downloading and executing a package from the NPM registry, which is a standard distribution platform for Node.js tools.
Audit Metadata